fix: remove fork(0) limitation from JMH runner

[not-matthias]

Summary

• Removes the workaround that forced forks=0 and warmupForks=0 when CodSpeed instrumentation is active
• Process filtering (COD-1788) is now supported, so benchmarks can safely run with user-configured fork counts
• Removes the CODSPEED_JVM_ALLOW_FORKS env var check (no longer needed)

Resolves COD-2447

[linear]

COD-2447 Remove @Fork(0) limitation from JMH

What?

How?

[codspeed-hq]

Merging this PR will degrade performance by 42.31%

⚡ 17 improved benchmarks
❌ 18 regressed benchmarks
✅ 90 untouched benchmarks
⏩ 9 skipped benchmarks¹

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Benchmark	BASE	HEAD	Efficiency
❌	findHighestSetBit[1024]	30 ns	34 ns	-11.76%
⚡	nQueensSolver[4]	1.6 µs	1.4 µs	+15.64%
⚡	nQueensSolver[5]	10 µs	8.7 µs	+14.72%
⚡	nQueensSolver[6]	16.1 µs	14.4 µs	+11.82%
⚡	permutations[5]	14.4 µs	13.1 µs	+10.17%
⚡	nQueensSolver[8]	519.1 µs	448.7 µs	+15.68%
❌	reverseBits[0]	29 ns	41 ns	-29.27%
❌	reverseBits[255]	29 ns	41 ns	-29.27%
⚡	coinChange[50]	733 ns	635 ns	+15.43%
❌	reverseBits[65535]	29 ns	41 ns	-29.27%
⚡	coinChange[200]	3.2 µs	2.8 µs	+13.39%
❌	reverseBits[1024]	29 ns	41 ns	-29.27%
❌	reverseBits[42]	29 ns	41 ns	-29.27%
⚡	coinChange[100]	1.6 µs	1.4 µs	+14.78%
⚡	fibonacciBottomUp[10]	544 ns	483 ns	+12.63%
❌	longestIncreasingSubsequence[10]	90 ns	100 ns	-10%
⚡	levenshteinDistance[saturday sunday]	791 ns	404 ns	+95.79%
❌	fibonacciMemo[10]	25 ns	28 ns	-10.71%
❌	fibonacciOptimized[20]	15 ns	26 ns	-42.31%
❌	fibonacciOptimized[40]	26 ns	37 ns	-29.73%
...	...	...	...	...

ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.

---

_{Comparing cod-2447-remove-fork0-limitation-from-jmh (4b7a767) with main (b43be0b)}

Footnotes

1. 9 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩